Information Technology Reference
In-Depth Information
Often, increasingly deep analysis drives the starting point back as we realize that
earlier occurrences were relevant.
It is usually first necessary to clarify the referential structure of the chat postings
and how they relate to events in the whiteboard or to the comings and goings of
participants. The threading of the chat postings provides the primary structure of
the online, text-based discourse in much the same way that turn taking provides
the core structure of spoken informal conversation. Because of the overlap in the
typing of chat postings, it is sometimes tricky to figure out who is responding to
what. Looking at the timestamps of posts and even at the timestamps of awareness
messages about who is typing can provide evidence about what was visible when a
posting was being typed. This can often suggest that a given post could or could not
have been responding to a specific other post, although this is sometimes impossible
to determine. When it is hard for the analyst to know the threading, it may have also
been hard for most of the chat participants (other than the typist) to know; this may
result in signs of trouble or misunderstandings in the subsequent chat.
The test of
correctness
of chat interaction analysis is not a matter of what was
in individuals' minds but of how postings function in the interaction. Most of the
multi-layered referencing takes place without conscious awareness by the partic-
ipants, who are experts at semantic, syntactic and pragmatic referencing and can
design utterances in response to local resources without formulating explicit plans
(Suchman, 2007). Thus, inspection of participants' memories would not reveal
causes. Of course, participants could retroactively tell stories about why they posted
what they did, but these stories would be based upon their current (not original)
interpretations using their linguistic competence and upon their response to their
current (not original) situation, including their sense of what the person interview-
ing them wants to hear. Thus, interpretations by the participants are not in principle
privileged over those of the analyst and others with the relevant interpretive compe-
tence (Gadamer, 1960/1988). The conscious memories that a participant may have
of the interaction are, according to Vygotsky's theory, just more interaction—but
this time sub-vocal self-talk; if they were brought into the analysis, they would be
in need of interpretation just as much as the original discourse.
Since our research question involves the group as the unit of analysis, we do not
raise questions in the data session about what one student or another may have been
doing, thinking or feeling as an individual. Rather, we ask what a given posting
is doing interactionally within the group process, how it responds to and takes up
other posts and what opportunities it opens for future posts. We look at how a post
is situated in the sequential structure of the group discourse, in the evolving social
order and in the team's meaning making. What is this posting doing here and now
in the referential network? Why is it “designed to be read” (Livingston, 1995) in
just this way? How else could it have been phrased and why would that not have
achieved the same effect in the group discourse?
We also look at how a given posting
positions
(Harré & Moghaddam, 2003) both
the author and the readers in certain ways. We do not attribute constant personali-
ties or fixed roles to the individuals, but rather look at how the group is organized
through the details of the discourse. Perhaps directing a question toward another
